I love statistics, but I am just not very good at it, and find much of it extremely counter intuitive (which is why it is ‘fun’). The Monty Hall problem floored me, but then Paul Erdos got it wrong too (I am told), so I am in good — and numerate — company. During my intercalated degree in addition to a research methods tutorials (class size, n=2), we had one three hour stats practical each week (class size, n=10). We each used a Texas calculator, and working out a SD demanded concentration. Never mind, that during the rest of the week we were learning how to use FORTRAN and SPSS on a mainframe, ‘slowing’ down the process was useful.
Medicine has big problems with statistics although it is often not so much to do with ‘mathematical’ statistics but evidence in a broader sense. IMHO the biggest abusers are the epidemiologists and the EBM merchants with their clickbait NNT and the like. But I do think this whole field deserves much greater attention in undergraduate education, and cannot help but feel that you need much more small group teaching over a considerable period of time. Otherwise, it just degenerates into ‘What is this test for?’ exam fodder style of learning.
The problems we have within both medicine and medical research have been talked about for a long while. Perhaps things are improving, but it is only more recently that this topic has been acknowledged as a problem amongst practising scientists (rather than medics). This topic certainly resurfaces with increased frequency, and there have been letters on it in Nature recently. I like this one:
Too many practitioners who discuss the misuse of statistics in science propose technical remedies to a problem that is essentially social, cultural and ethical (see J. Leek et al. Nature 551, 557–559; 2017). In our view, technical fixes are doomed. As Steven Goodman writes in the article, there is nothing technically wrong with P values. But even when they are correct and appropriate, they can be misunderstood, misrepresented and misused — often in the haste to serve publication and career. P values should instead serve as a check on the quality of evidence.
I think you could argue with the final sentence of this (selected) quote, but they are right about the big picture: narrow technical solutions are not the problem here. Instead, we are looking at a predictable outcome of the corruption of what being a scientist means.
The Osborne effect is described in Wikipedia as follows:
The Osborne effect is a term referring to the unintended consequences of a company announcing a future product, unaware of the risks involved or when the timing is misjudged, which ends up having a negative impact on the sales of the current product. This is often the case when a product is announced too long before its actual availability. This has the immediate effect of customers canceling or deferring orders for the current product, knowing that it will soon be obsolete, and any unexpected delays often means the new product comes to be perceived as vaporware, damaging the company’s credibility and profitability.
AI and associated technologies will have major effects in some areas of medicine. Think skin cancer diagnosis, for certain; or this weekend story in the FT on eye disease; and radiology and pathology. This then begs the question, whether these skills are so central to expertise within a clinical domain, that students should think hard about these areas as a career. Of course, diagnosis of skin lesions is not all a clinical expert in this domain does. Ditto, ophthalmologists do more than look at retinas. Automated ECG readers have not put cardiologists out of work, after all. And many technical advances increase — not reduce — workloads.
But at some stage, people might want to start wondering if some areas of medicine are (not) going to be secure as long term careers. The Osborne metaphor should be a warning about how messy all this could be. Hype, has costs.
The surge in open-access predatory journals is making it harder for contributors and readers to distinguish these from legitimate publications — a confusion that is fostered by the predatory-journal industry. One solution could be to deploy a variant of a well-established quality-control test. The scientific community could submit replicate test articles several times a year to a wide array of open-access journals, suspect and non-suspect.
From Steven N Goodman who, as ever, is worth reading. Of course, in one sense, it is a question of serial monogamy, or polygamy.
Not the usual stuff Nobel Laureates spiel, but take a look. An article in Quartz is here and Wikipedia is useful on him
I like the ‘biogibberish’ epithet. And cannot help but suspect he would agree with David Hubel’s line that reading most papers now is like chewing sawdust. But you can see the fire still burns: you have to be dissatisfied with the state of the universe. How polite or angry you are is a question of personal style.
As each year passes, the once celebrated barriers between man and the other animals become less secure. Once we were the only tool makers, once we were the ones who discovered drugs or used technology. This report is about how finches commandeer cigarette butts for a new purpose.
That idea has been around, though never proved, since 2012. This was when Dr Suárez-Rodríguez showed that nests which had butts woven into them were less likely to contain bloodsucking parasites than were nests that did not. What she was unable to show was whether the nest-builders were collecting discarded cigarettes deliberately for their parasite-repelling properties, or whether that parasite protection was an accidental consequence of butts being a reasonably abundant building material.
And finches, again! Where would biology be without Darwin’s finches?
David Hubel, on statistics: “We could hardly get excited about an effect so feeble as to require statistics for its demonstration.”
I came across this (below), in my end of year clear out. And even if this was 2016, rather than 2017, it is as good a thought to open 2018 with, as any other. It is from a review of “Life’s Greatest Secret: The Race to Crack the Genetic Code”, by Matthew Cobb. The review is by H Allen Orr. NYRB
Finally, and perhaps most important, Life’s Greatest Secret highlights the power of the beautiful experiment in science. Though Cobb pays less attention to this subject than he might have, the period of scientific history that he surveys was the golden age of the beautiful experiment in biology. Biologists of the time—including Nirenberg with his UUU, Crick and Brenner with their triplet code work, and others including Matthew Meselson, Franklin Stahl, and Joshua Lederberg—were masters of the sort of experiment that, through some breathtakingly simple manipulation, allowed a decisive or nearly decisive solution to what previously seemed a hopelessly complex problem. Such experiments represent a species of intellectual art that is little appreciated outside a narrow circle of scientists……..
But the larger lesson of Life’s Greatest Secret is one that may be worth remembering. When scientists require definitive answers, not merely suggestive patterns, they require experiments that are decisive and, if all goes well, beautiful.
“For example, I studied Physics, so I learned about how physicists think… and it is not how most people think. They have these tricks which turn difficult problems into far easier problems. The main lesson I took away from Physics is that you can often take an impossibly hard problem and simply represent it differently. By doing so, you turn something that would take forever to solve into something that is accessible to smart teenagers.”
But the opposite is now much more common. I think there are whole swathes of modern institutional and corporate life, that are designed to make the simple, complicated. At best, simple may sometimes be wrong, but complicated is usually useless — or much worse. I seem to remember Paul Jannsen, when asked why we do not seem to be able to discover revolutionary new drugs like we once did, respond: ‘in those days the idea of obviousness still existed’.
This is a term I first learned from Clark Glamour and colleagues in Android Epistemology. Dermofit was a failed attempt to try and invent such a prosthesis.
Thinkers and thinking societies build tools that enhance their own thinking. When the speed of the positive feedback increases rapidly, we see a scientific and cultural revolution. When grit is put into the cogs or the base metals diluted, the opposite happens.
Last week I was giving a talk about tech, medicine and medical education, and for the life of me could not remember the following example, showing how key representation is to our intellectual toolbox. Worse, I knew it had an Edinburgh connection. Wikipedia has more.
This is from an article in Nature. And the problem is resolving differences in experimental results between labs.
But subtle disparities were endless. In one particularly painful teleconference, we spent an hour debating the proper procedure for picking up worms and placing them on new agar plates. Some batches of worms lived a full day longer with gentler technicians. Because a worm’s lifespan is only about 20 days, this is a big deal. Hundreds of e-mails and many teleconferences later, we converged on a technique but still had a stupendous three-day difference in lifespan between labs. The problem, it turned out, was notation — one lab determined age on the basis of when an egg hatched, others on when it was laid.
Now my title is from Blake:
He who would do good to another must do it in Minute Particulars: general Good is the plea of the scoundrel, hypocrite, and flatterer, for Art and Science cannot exist but in minutely organized Particulars.
And yet, I think I am using the quote in a way he would have strongly disagreed with. Some of the time ‘Minute particulars’ are not the place to be if you want to change the world. Especially in biology.
Little evidence is found for higher-order organization into 30- or 120-nm fibers, as would be expected from the classic textbook models based on in vitro visualization of non-native chromatin.
Well, chromatin structure might not be everybody’s cup of tea but I once shared an ‘office’ with a couple of French / Polish researchers in Strasbourg. It was all above my head, so I had to make do with the textbooks, and I stuck to my simple cloning of upstream regulatory regions of a retinoid receptor. Now, it appears from this article in Science, the textbooks will need rewriting. Science works.
Hours later, with Blackburn’s approval, the institute issued comments on the scientific records of the two women. It had “invested millions of dollars” in each scientist, Salk stated, but a “rigorous analysis” showed each “consistently ranking below her peers in producing high quality research and attracting” grants. Neither has published in Cell, Nature, or Science in the last 10 years, it said. Lundblad’s salary “is well above the median for Salk full professors ($250,000) … yet her performance has long remained within the bottom quartile of her peers.” The institute wrote that Jones’s salary, in the low $200,000 range, “aligns” with salaries at top universities, although she “has long remained within the bottom quartile of her peers.”
This is from an article in Science (Gender discrimination lawsuit at Salk ignites controversy). The context is a sex discrimination case, but the account is about an astonishing lack of vision. Short termism of stock markets is not the only way value is being destroyed by a cash-in-hand mentality. The best rugby coaches have rarely been the greatest players. Nobel laureates may not be the best leaders. No (bull) shit here…
Terrific interview with Alan Kay. Familiar memes, but I do not tire of them.
The business of a university is to help students learn contexts that they were unaware of when they were in high school.
His use of the word context encompasses intellectual creations such as reading, writing, printing etc. His oft quoted quip: a change in context is worth 80 IQ points
Or so says an article in Nature. No we don’t , is my response
Philanthropists are flying blind because little is known about how to donate money well. Facebook co-founder Mark Zuckerberg’s US$100-million gift to schools in Newark, New Jersey, reportedly achieved nothing. Some grants to academic scientists create so much administration that researchers are better off without them. And some funders’ decisions seem to be no better than if awardees were chosen at random, with the funded work achieving no more than the rejecte
There is no science to philanthropy. You can study it, you can come up with ideas about it, and try to meld systems of rationality about it. But this is just an abuse of the word science, an abuse meant to demarcate this area of activity from things that are non-science and are, by implication, less robust or rigorous. This is one of the ways the science (and STEM) lobby misunderstand the world. But, the quoted paragraph, does of course say something meaningful.
Institutions with histories matter. It is just that in many instances innovation often comes from the periphery. I think this is often true in many fields: science, music, even medical education. It is not always this way, but often enough to make me suspicious of the ‘centre’. The centre of course gets to write the history books.
An article by Mark Mazower in the NYRB, praising Richard Evans, the historian of the Third Reich, caught my attention. It seems that nobody in the centre was too excited about understanding the event that changed much of the world forever. Mazower writes:
If you wanted to do research on Saint Anselm or Cromwell, there were numerous supervisors to choose from at leading universities; if you wanted to write about Erich Ludendorff or Hitler, there was almost no one. The study of modern Europe was a backwater, dominated by historians with good wartime records and helpful Whitehall connections—old Bletchley Park hands and former intelligence officials, some of whom had broken off university careers to take part in the war and then returned.
Forward-looking, encouraging of the social sciences, open to international scholarship from the moment of its establishment, St. Antony’s is the college famously written off by the snobbish Roddy Martindale in John le Carré’s Tinker, Tailor, Soldier, Spy as “redbrick.” The truth is that it was indeed the redbrick universities, the creations of the 1950s and 1960s, that gave Evans and others their chance and shaped historical consciousness as a result. The Evans generation, if we can call them that, men (and only a very few women) born between 1943 and 1950, came mostly from the English provinces and usually got their first jobs in the provinces, too.
It is interesting how academics who had had career breaks were important. And how you often will need new institutions to change accepted practice. All those boffins whose careers were interrupted by the war led to the flowering of invention we saw after the second world war. You have to continually recreate new types of ivory towers. But I see little of this today. Instead, we live in an age of optimisation, rather than of optimism that things can be different. The future is being captured by the present ever more than it once was. At least in much of the academy.
Spectral authors also haunt the scientific canon. One physicist, frustrated at having his paper repeatedly rejected, finally saw it published after changing the title and adding a fictitious co-author, Stronzo Bestiale. It means “total asshole” in Italian.
Seriously, if you suggested the world we have now of predatory journals and the tyranny of metrics, would any sane scientist in 1960 think it possible? Uncle Syd once remarked that people no longer read papers they just xeroxed them. Now we do not even do that: metadata is all.
I thought I would have read this before, but maybe I put it to one side and foolishly forgot. It is a fitting description of Jacob Bronowksi by his wife, Rita. One thing — amongst many — caught my eye.
As a very young man he would travel miles every week to outlying villages in England to give what were called Workers’ Educational Association lectures. Quite literally he would travel through snow and fog to village halls to speak to 8 or 10 people who had equally braved the elements. I sometimes would think it a pity there were not hundreds thereto hear him. Little did I imagine that with radio and then television he would in fact finally reach millions.
And I would respond: you have to want to learn, and you have to want to educate.
But I can’t stop here. One bit of the jigsaw I didn’t know:
After receiving his Ph.D. and conducting 3 years of research, it became clear that, being a Jew, Bruno would not be made a Fellow of his college (Jesus College, Cambridge). He decided to ‘drop out’. Like so many young students (hippies, 30 years later), bearded and down-at-heel, he went to Paris to write. There he met, among others, Samuel Beckett, and they jointly edited an anthology called European Caravan..
It ends with his own words
What makes the biological machinery of man so powerful is that it modifies his actions through his imagination. It makes him able to symbolize, to project himself into the consequences of his acts, to conceptualize his plans, to weigh them, one against another, as a system of values… We, as men, are unique. We are the social solitaries … We are the creatures who have to create values in order to elucidate our own conduct, so that we learn from it and can direct it into the future (emphasis, mine)
In LEONARDO, Vol. 18, No. 4, pp. 223-225,1985
Q: What’s at stake when scientists fib?
A: Science is the last institution where being honest is a quintessential part of what you’re doing. You can do banking and cheat, and you’ll make more money, and that money will still buy you the fast cars and the yachts. If you cheat in science, you’re not making more facts, you’re producing nonfacts, and that is not science. Science still has this chance of giving a lead to democratic societies because scientific values overlap strongly with democratic values.
Interview with Harry Collins about his book: Gravity’s Kiss: The Detection of Gravitational Waves Harry Collins MIT Press, 2017. 414 pp.
Bruce Alberts talks a lot of sense about science education and education in general. And of course he produced a book that ‘educated’ a whole generation (or more) of people like me. But in this recent Science piece he is taking on some of the big questions, questions that have been asked before, but for which few have managed to follow through on. As ever, the emphases are mine.
In previous commentaries on this page, I have argued that “less is more” in science education, and that learning how to think like a scientist—with an insistence on using evidence and logic for decision-making—should become the central goal of all science educators. I have also pointed out that, because introductory science courses taught at universities define what is meant by “science education,” college science faculty are the rate-limiting factor for dramatically improving science education at lower levels.
For example, there is a long-standing belief that every introductory college biology course must “cover” a staggering amount of knowledge. There is no time to focus on a much more important goal—insisting that every student understand exactly how scientific knowledge is generated. Science is not a belief system; it is, instead, a very special way of learning about the true nature of the observable world.
His phrase, “college science faculty are the rate-limiting factor for dramatically improving science education at lower levels”, could equally apply to medicine and medical teachers. It is not hyperbole to say these are some of the central problems of our time. And it is not just science education that is the issue.
Universities are idea factories. Current corporatization approaches emphasize the factory rather than the ideas.
Ralf Buckley in Nature. I would say— for the short term at least — unless somebody finds a way to create new ‘dissenting academies’ things in UK higher ed will get worse.
As if by a miracle, once up and running, the Mark 1 telescope was the only instrument that could both detect the first Soviet and American satellites and transmit instructions to them. Amazing as it now seems, the need for such a telescope had escaped both the telecommunications industry and the military leaders of both superpowers.
Despite its spectacular success, which included tracking the Sputnik 1 satellite mission in 1957, the government did nothing to alter the remaining debt, being bound by the iron restraint of Treasury rules. It was Lord Nuffield who did so, thereby demonstrating the superiority of aristocratic, rather than state support, to science – and indeed to all intellectual activity, a view which Lovell expressed frequently and forcefully to the end of his life.
Sir Fred Hoyle’ obituary of Sir Bernard Lovell. I fear Hoyle is right — at least if we realise we need more Fred Hoyles. Now, they are not aristocrats, but US philanthropists.
Interesting editorial in Nature. And unexpected. The issue is support for science and the state of politics in the US.
Just telling the same old stories won’t cut it. The most seductive of these stories — and certainly the one that scientists like to tell themselves and each other — is the simple narrative that investment in research feeds innovation and promotes economic growth. ‘It’s the economy, stupid’, so the saying goes, and as nations become a little less stupid by pushing against the frontiers of knowledge, so the benefits of all this new insight spread from the laboratory to the wider population, as improvements in the standard of living and quality of life. This comfortable story has all the hallmarks of a bubble waiting to pop.
The article goes on:
It is right that more scientists should tell stories of the good their research can do. But it is more important and urgent than ever that researchers should question how these stories really end — and whether too many of the people they claim to act for don’t really get to live happily ever after.
Much science is in a vacuous bubble, and arguments for more funding from its practitioners is increasingly viewed as self serving. Universities share some or much of this blame, all too happy to ‘shift more units’. This lack of intellectual honesty will harm academia in the long term. The one uniting feature that justifies higher education is the pursuit of truth in whichever direction enquiry moves. Universities are not businesses, profit centres, or corporations. They have a different set of norms that are distinct from those advertised by much of the rest of the corporate world (or government). STEM has never been enough, and truthfulness is not something you can opt in or out of, like you can some undergraduate modules. The role for universities — and science — is greater than ever: the issue is whether the universities have the necessary leadership. Even with the right leaders, it is a tough ask.
In 1924, a 30-year old journalist on the Daily Express came to Cambridge aiming to interview Haldane. She was Charlotte Burghes, née Franken, and she had a young son, Ronnie. Haldane and Charlotte became lovers, but before they could marry she had to seek a divorce, a procedure that carried substantial social stigma at that time. A university committee resolved to strip Haldane of his readership, which was only restored by successful legal action.
There are many worthwhile insights on show in the THE interview with the Nobel physicist Saul Perlmutter, ‘You can’t order up a breakthrough’. But this caught me eye:
“I think for students it’s never too early…to realise that they should be helping…to figure out the world together, not just learning the received facts and the things we already know,” he says. “We find ourselves looking at a world where I don’t think almost any of the problems I see today would worry me, if we knew how to work together and how to think through problems together in a rational way that wove together fears and needs with a rational understanding of the world…Maybe one of the best ways into that is to start teaching.” (emphasis mine)
I think this is the kernel of the problem we face, and the trite “we need more STEM’ or ‘teach all students to code’, is missing the key issue. We are, as has been said before, a ‘civilisation that is face to face with its own implications”.
In June 2016, data released by the UK’s Office for National Statistics revealed that there had been 52,400 more deaths in the year following June 2015 compared with the same period a year before; an annual rise of 9%. These rises in mortality rates are unprecedented in post-war times. In England and Wales, the increase was almost entirely in the population aged over 55 years, predominantly in those aged over 75, and was largely attributed to dementia and Alzheimer’s disease, with influenza being suggested as a minor contributory factor. It was mostly those with long-term care needs who were dying earlier.
In addition, in late 2016, official data was released for Scotland showing no rise in life expectancy for men and women for the first time in 160 years.
These are two quotes from an article in yet another new journal, Nature Human Behaviour. The title of the piece, ‘Policymakers should not act like scientists’, is worth thinking over. There is always a bias built into ‘tests of departure from..’ and those who control the funding, can control what counts as legitimate evidence. And the null hypothesis of the ‘classical statistical paradigm’ is not appropriate for many types of problems. From somewhere, I remember a comment from Paul Jannsen along the lines of, ‘in those days the idea of obviousness still existed’. As we saw last week, in the debate about the NHS, in the mouths of politicians a handful can mean upwards of 10,000. As is often misquoted: the plural of anecdote is data.
From an article in Nature describing how the US biomedical workforce has changed over recent times.
Our analysis of IPUMS-USA data reveals a cohort that entered the laboratory workforce as NIH funding grew from US$13.7 billion in 1998 to $28.1 billion in 2004. These ‘doubling boomers’ arguably suffered most as funds subsequently decreased (when adjusted for inflation). In 2004, there were nearly 26,000 individuals under 40 with PhDs working as biomedical scientists. By 2011, there were nearly 36,000. Over this period, the number of faculty jobs did not increase. Indeed, the number of openings expected as a result of academics retiring has declined since 1995, when federal law made it illegal for universities to mandate retirement at age 65 (ref. 3).
The work environment that this cohort faces is unlike anything seen before, despite previous booms and busts4. Today in the United States, four out of five PhD biomedical researchers work outside academia — a record high (see ‘Lab labour’). They earn, on average, almost $30,000 more a year than their academic counterparts, and feel less pressure to produce scientific publications.
There are some obvious points. The doubling was crazy at the time (some of us said that, then), and even more so in hindsight. Universities rushed for the gold, with little wider thought. Second, careers are the ‘long now’ and getting longer. Personal investment relies on a certain degree of continuity and stability, and there will be a hangover, that the universities will now have to deal with. Finally, the obsession with growth by universities is dangerous. Haldane’s essay, ‘On being the right size’ comes to mind. Scaling matters, as does thinking about long term rather than short term success.
“Throughout her career, Gonzalez has done “a bit of everything” at LIGO, she says. For a while, she took on the crucial task of diagnosing the performance of the interferometers to make sure that they achieved unparalleled sensitivity — which is now enough to detect length changes in the 4-kilometre-long arms of the interferometers to within one part in 1021, roughly equivalent to the width of DNA compared with the orbit of Saturn. “
It ain’t biology, then. Nature
Interesting story in Nature highlighting instances where instead of doing post-docs, young biologists have raised funding to set up their own companies. Of course, most start-ups fail, but then most really interesting research projects should fail. Y Combinator is getting into this area, which surprised me. As the age of getting your first grant gets higher, and with the increasingly dysfunctional nature of much academic (medical) science, the attractions are obvious. I was sceptical (and still am) that the ‘software’ model would work in this area.
Cartoon characters not infrequently run off the edge of a cliff. Pause. They then realise there is nothing there to support their running. Time lags are awkward to deal with in any analysis, but since most things do not happen overnight, they are ubiquitous. In analysis, we replace with a fudge factor. Or we ignore them.
I haven’t seen much comment on what I think is the most interesting aspect of the science news over the last few weeks. Here is a line from THE.
Five out of nine laureates in the core prizes for physics, chemistry, medicine and economic sciences were born in the UK. All crossed the pond as rather valuable immigrants to the US.
UK science, and many UK universities, have been in profit-harvesting mode for a long time now. Over the edge of that cliff. Things are going to
fall apart. OK, what the hell:
The falcon cannot hear the falconer;
Things fall apart; the centre cannot hold;
Knowledge is not a looseleaf notebook of facts. Above all, it is a responsibility for the integrity of what we are, primarily of what we are as ethical creatures. You cannot possibly maintain that it informed integrity if you let other people run the world for you what you yourself continue to live out of a rag bag of morals that come from past beliefs. That is really crucial today. You can see it is pointless to advise people to learn differential equations, or to do a course in electronics or in computer programming. And yet, 50 years from now, if an understanding of man’s origins, his evolution, his history, his progress is not the commonplace of the schoolbooks, we shall not exist. The commonplace of the schoolbooks of tomorrow is the adventure of today, and that is what we are engaged in……
It sounds very pessimistic to talk about western civilisation with a sense of retreat. I’ve been so optimistic about the ascent of man; and I going to give up at this moment? Of course not. The ascent of man will go on. But do not assume that it will go on carried by Western civilisation as we know it. We are being weighed in the balance at this moment.
The Ascent of Man, Jacob Bronowksi 1973.